The Mess of Crisis Time LIBOR
Peter Tchir gives some interesting insight into the LIBOR vs CDS prices.
Here are the 3 month USD LIBOR submissions for some of the banks, along with where 1 year CDS was quoted at the time. The 1 year CDS data is not the best, but is indicative and certainly reflects what I remember as the relative safety perception. I’m also looking into CD rates, bonds, and stock prices, but I find a few things interesting here.
| Date | JPM | Citi | BofA | DB | HBOS | RBS | Barc | UBS | CS | Nor | BOTM | HSBC |
| 10/3/2008 | 100 | 345 | 105 | 114 | 228 | 235 | 189 |
171 |
91 | 77 | 55 | 52 |
| 10/3/2008 | 4.10 | 4.02 | 4.50 | 4.05 | 4.90 | 4.50 | 5.00 | 4.35 | 4.50 | 4.40 | 4.60 | 4.10 |
| 10/9/2008 | 103 | 400 | 113 | 102 | 181 | 246 | 149 | 191 | 83 | 104 | 78 | 45 |
| 10/9/2008 | 4.40 | 4.45 | 4.80 | 4.65 | 5.00 | 5.00 | 5.10 | 4.75 | 5.00 | 4.75 | 4.95 | 4.30 |
| 10/31/2008 | 68 | 174 | 81 | 95 | 113 | 98 |
110 |
108 | 104 | 137 | 65 | 68 |
| 10/31/2008 | 2.75 | 3.00 | 2.80 | 3.05 | 3.10 | 3.00 | 3.20 | 3.02 | 3.05 | 3.05 | 3.20 | 3.00 |
| 11/20/2008 | 98 | 369 | 104 | 112 | 146 | 143 | 156 | 146 | 125 | 94 | 98 | |
| 11/20/2008 | 1.85 | 2.10 | 2.15 | 2.20 | 2.15 | 2.23 | 2.40 | 2.14 | 2.25 | 2.18 | 2.20 | 2.10 |
Citi stands out in that their CDS was consistently flashing a warning signal, yet their LIBOR submissions were in the middle to low end of the range. On the 9th of October, they seem very low (in fact they weren’t included in the calculation as they were so low), yet their CDS is the highest. They consistently seem to have spreads that at the very least seem to warrant further investigation. Had they posted 5% like many other banks rather than 4.45%, LIBOR would have set at 4.806% instead of 4.75%. That is 5 bps. Maybe the Citi rates make sense, but if you are worried that someone was setting artificially low rates, they catch my eye.
RBS strikes me as a bit low relative to Barclay’s on occasion. October 3rd seems particularly low, relative to both Barclay’s and HBOS.
UBS is another one where they are consistently lower on the LIBOR side than CS, yet in many instances their CDS would indicate they should be wide.
If on October 3rd, Citi posted 4.5% just like BofA, and RBS posted 5% just like Barclays, and UBS posted 4.5% just like CS , then LIBOR would have been 4.403% instead of 4.334%. That is about 7 bps. You are starting to talk about some serious numbers.
I am not saying their numbers are off, and maybe they are easily justified, but if I was focusing resources on where LIBOR may have been really pushed from true value I would look there.
I used a nice yellow to highlight levels that seemed conservative. For all the grief Barclay’s is taking, they are consistent and I find their levels easier to justify than some of the others.
Someone Didn’t Get the Message that Lehman Filed for Bankruptcy
| Date | JPM | Citi | BofA | Lloyds | HSBC | RBS | HBOS | Barc | DB | West | UBS | CS |
| 9/12 | 2.80 | 2.82 | 2.78 | 2.80 | 2.79 | 2.82 | 2.83 | 2.84 | 2.82 | 2.83 | 2.82 | 2.85 |
| 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | -0.01 | 0.00 | 0.00 | 0.00 | 0.00 | 0.03 | |
| 9/15 | 2.79 | 2.81 | 2.78 | 2.83 | 2.78 | 2.81 | 2.82 | 2.83 | 2.85 | 2.82 | 2.82 | 2.85 |
| -0.01 | 0.00 | 0.00 | 0.03 | -0.01 | 0.00 | -0.01 | -0.01 | 0.03 | -0.01 | 0.00 | 0.00 | |
| 9/16 | 2.82 | 2.85 | 2.85 | 2.83 | 2.90 | 2.85 | 2.85 | 3.00 | 3.00 | 2.87 | 2.88 | 3.00 |
| 0.03 | 0.04 | 0.07 | 0.00 | 0.12 | 0.04 | 0.03 | 0.17 | 0.15 | 0.05 | 0.06 | 0.15 | |
| 9/17 | 3.03 | 2.99 | 3.00 | 2.85 | 3.05 | 3.05 | 3.07 | 3.20 | 3.10 | 3.08 | 3.05 | 3.10 |
| 0.21 | 0.14 | 0.15 | 0.02 | 0.15 | 0.20 | 0.22 | 0.20 | 0.10 | 0.21 | 0.17 | 0.10 | |
| 9/18 | 3.00 | 3.15 | 3.20 | 3.17 | 3.15 | 3.16 | 3.20 | 3.75 | 3.30 | 3.25 | 3.20 | 3.25 |
| -0.03 | 0.16 | 0.20 | 0.32 | 0.10 | 0.11 | 0.13 | 0.55 | 0.20 | 0.17 | 0.15 | 0.15 | |
| 9/19 | 3.00 | 3.20 | 3.20 | 3.19 | 3.22 | 3.20 | 3.20 | 3.75 | 3.20 | 3.21 | 3.20 | 3.25 |
| 0.00 | 0.05 | 0.00 | 0.02 | 0.07 | 0.04 | 0.00 | 0.00 | -0.10 | -0.04 | 0.00 | 0.00 | |
| 9/22 | 3.05 | 3.18 | 3.10 | 3.18 | 3.18 | 3.18 | 3.18 | 3.85 | 3.35 | 3.20 | 3.18 | 3.30 |
| 0.05 | -0.02 | -0.10 | -0.01 | -0.04 | -0.02 | -0.02 | 0.10 | 0.15 | -0.01 | -0.02 | 0.05 | |
| Full Chg | 0.25 | 0.37 | 0.32 | 0.38 | 0.39 | 0.37 | 0.34 | 1.01 | 0.53 | 0.37 | 0.36 | 0.48 |
It seems that funding didn’t get difficult going into the weekend and that on the first day no one thought it made a difference.
But then on the 16th, HSBC, Barc, DB and CS all raised their LIBOR by about 15 bps on average. Lloyd’s was seemingly oblivious.
The next day virtually everyone raised their rates, except for the still sleep Lloyd’s (or maybe they weren’t affected). At a quick glance, banks that had been “unaffected” on the 16th seemed to have the biggest moves on the 17th. Does that mean they didn’t experience it on the 16th, or just had been “slow” to update their systems? Given the amount of catch up the next day, I would check the 16th as a potential understatement, mostly because so many did it.
Lloyd’s was so low, that even catching up might not have impacted LIBOR as they were getting excluded, but still worth a look.
By the 18th, Barclay’s had separated itself marvelously from the pack. They were over 1% wider by the 22nd versus the 11th. No other bank came close. Since the complaint is that Barclay’s was submitting rates that were too low (or maybe that is later after the BofE “suggested” they get in line with other banks), it is worth checking many of these submissions and see if they reflected reality. The S&P was reasonably stable, so maybe the submissions are correct, but the noise level seems pretty high.
Anyways, it isn’t worth spending much time on pre-crisis levels. The moves were all too similar and without widespread collusion no one could move LIBOR around much. This is where there are some real potential to find problems. Honest mistakes, actual supportable levels, or something else would take time and effort to prove, but times of stress are where to look for outliers and patterns.